pandas groupby函数

pandas groupby函数

---

pandas的groupby函数一般会配合合计函数使用，比如，count、avg等等。

首先讲解几种模式，然后示例上场：

第一种：df.groupby(col)，返回一个按列进行分组的groupby对象；

第二种：df.groupby([col1,col2])，返回一个按多列进行分组的groupby对象；

第三种：df.groupby(col1)[col2]或者df[col2].groupby(col1)，两者含义相同，返回按列col1进行分组后，col2的值；

---

先创建一个DataFrame对象df：

import pandasas pd

from dateutil.parserimport parse

import datetimeas dt

import matplotlib.pyplotas plt

df= pd.DataFrame({'Date': ['2015-05-08','2015-05-07','2015-05-06','2015-05-05','2015-05-08','2015-05-07','2015-05-06','2015-05-05'],

'Sym': ['aapl','aapl','aapl','aapl','aaww','aaww','aaww','aaww'],

'Data2': [11,8,10,15,110,60,100,40],

'Data3': [5,8,6,1,50,100,60,120]})

创建一个DataFrame对象df 结果数据

---

第一种：df.groupby(col)，返回一个按列进行分组的groupby对象；

对日期进行分组，打印出结果样式，结果如下：

import pandasas pd

from dateutil.parserimport parse

import datetimeas dt

import matplotlib.pyplotas plt

df= pd.DataFrame({'Date': ['2015-05-08','2015-05-07','2015-05-06','2015-05-05','2015-05-08','2015-05-07','2015-05-06','2015-05-05'],

'Sym': ['aapl','aapl','aapl','aapl','aapl','aapl','aaww','aaww'],

'Data2': [11,8,10,15,110,60,100,40],

'Data3': [5,8,6,1,50,100,60,120]})

df1=df.groupby(df['Date'])

print(list(df1))

df.groupby(col)

第二种：df.groupby([col1,col2])，返回一个按多列进行分组的groupby对象；

import pandasas pd

from dateutil.parserimport parse

import datetimeas dt

import matplotlib.pyplotas plt

df= pd.DataFrame({'Date': ['2015-05-08','2015-05-07','2015-05-06','2015-05-05','2015-05-08','2015-05-07','2015-05-06','2015-05-05'],

'Sym': ['aapl','aapl','aapl','aapl','aapl','aapl','aaww','aaww'],

'Data2': [11,8,10,15,110,60,100,40],

'Data3': [5,8,6,1,50,100,60,120]})

df1=df.groupby(['Date','Sym'])

print(list(df1))

df.groupby([col1,col2])代码 df.groupby([col1,col2])结果

---

第三种：df.groupby(col1)[col2]或者df[col2].groupby(col1)，两者含义相同，返回按列col1进行分组后，col2的值；

import pandasas pd

from dateutil.parserimport parse

import datetimeas dt

import matplotlib.pyplotas plt

df= pd.DataFrame({'Date': ['2015-05-08','2015-05-07','2015-05-06','2015-05-05','2015-05-08','2015-05-07','2015-05-06','2015-05-05'],

'Sym': ['aapl','aapl','aapl','aapl','aapl','aapl','aaww','aaww'],

'Data2': [11,8,10,15,110,60,100,40],

'Data3': [5,8,6,1,50,100,60,120]})

df1=df.groupby(df['Date'])['Sym']

print(list(df1))

---

---

df.groupby(col1)[col2]

---

df.groupby(col1)[col2]

---

import pandasas pd

from dateutil.parserimport parse

import datetimeas dt

import matplotlib.pyplotas plt

df= pd.DataFrame({'Date': ['2015-05-08','2015-05-07','2015-05-06','2015-05-05','2015-05-08','2015-05-07','2015-05-06','2015-05-05'],

'Sym': ['aapl','aapl','aapl','aapl','aapl','aapl','aaww','aaww'],

'Data2': [11,8,10,15,110,60,100,40],

'Data3': [5,8,6,1,50,100,60,120]})

df1=df['Sym'].groupby(df['Date'])

print(list(df1))

df['Sym'].groupby(df['Date']) df['Sym'].groupby(df['Date'])

---

好啦，就到这啦，谢谢